All of lore.kernel.org
 help / color / mirror / Atom feed
* real free space on btrfs volume (performance impact)
@ 2012-01-10 14:13 Michal Suba
  2012-01-10 18:40 ` Mitch Harder
  0 siblings, 1 reply; 3+ messages in thread
From: Michal Suba @ 2012-01-10 14:13 UTC (permalink / raw)
  To: linux-btrfs

Hello

  we are currently investigating performance issue on system runing 
above btrs filesystem. Is it possible, that performance is impacted by 
lack of free space? Also, how to get info about real free space on btrfs 
volume?

# btrfs-show /dev/sdb1
Label: opt  uuid: 28a55827-e677-47a9-98d5-d31eb3d71436
     Total devices 1 FS bytes used 167.83GB
     devid    1 size 240.00GB used *229.25GB* path /dev/sdb1

Btrfs Btrfs v0.19

# btrfs filesystem df /opt
Data: total=213.23GB, used=165.26GB
System, DUP: total=8.00MB, used=40.00KB
System: total=4.00MB, used=0.00
Metadata, DUP: total=8.00GB, used=2.57GB

# df -h /opt
Filesystem            Size  Used Avail Use% Mounted on
/dev/sdb1             240G  171G   59G  75% /opt

How come that there is difference detween btrfs-show and df .. 40GB Is 
the space really usead or can I claim it back? (there are no snapshots)

# btrfs subvolume list /opt
#

  Thanks
michal


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: real free space on btrfs volume (performance impact)
  2012-01-10 14:13 real free space on btrfs volume (performance impact) Michal Suba
@ 2012-01-10 18:40 ` Mitch Harder
  2012-01-10 18:54   ` cwillu
  0 siblings, 1 reply; 3+ messages in thread
From: Mitch Harder @ 2012-01-10 18:40 UTC (permalink / raw)
  To: Michal Suba; +Cc: linux-btrfs

2012/1/10 Michal Suba <michal.suba@pantheon.sk>:
> Hello
>
> =A0we are currently investigating performance issue on system runing =
above
> btrs filesystem. Is it possible, that performance is impacted by lack=
 of
> free space? Also, how to get info about real free space on btrfs volu=
me?
>
> # btrfs-show /dev/sdb1
> Label: opt =A0uuid: 28a55827-e677-47a9-98d5-d31eb3d71436
> =A0 =A0Total devices 1 FS bytes used 167.83GB
> =A0 =A0devid =A0 =A01 size 240.00GB used *229.25GB* path /dev/sdb1
>
> Btrfs Btrfs v0.19
>
> # btrfs filesystem df /opt
> Data: total=3D213.23GB, used=3D165.26GB
> System, DUP: total=3D8.00MB, used=3D40.00KB
> System: total=3D4.00MB, used=3D0.00
> Metadata, DUP: total=3D8.00GB, used=3D2.57GB
>
> # df -h /opt
> Filesystem =A0 =A0 =A0 =A0 =A0 =A0Size =A0Used Avail Use% Mounted on
> /dev/sdb1 =A0 =A0 =A0 =A0 =A0 =A0 240G =A0171G =A0 59G =A075% /opt
>
> How come that there is difference detween btrfs-show and df .. 40GB I=
s the
> space really usead or can I claim it back? (there are no snapshots)
>
> # btrfs subvolume list /opt
> #
>

The btrfs-show command is being deprecated.  It's output can be easy
to misunderstand, but it probably won't be corrected since it's going
away at some point.

Basically, what this is telling you is that 229.25GB is "committed"
(213.23GB Data + 2 x8.00GB Metadata (because it's duplicated) + 2 x
8.00MB System.

However, all the "committed" space is not being used (which is clearer
in the 'btrfs filesystem df' command).
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" =
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: real free space on btrfs volume (performance impact)
  2012-01-10 18:40 ` Mitch Harder
@ 2012-01-10 18:54   ` cwillu
  0 siblings, 0 replies; 3+ messages in thread
From: cwillu @ 2012-01-10 18:54 UTC (permalink / raw)
  To: Mitch Harder; +Cc: Michal Suba, linux-btrfs

On Tue, Jan 10, 2012 at 12:40 PM, Mitch Harder
<mitch.harder@sabayonlinux.org> wrote:
> 2012/1/10 Michal Suba <michal.suba@pantheon.sk>:
>> Hello
>>
>> =C2=A0we are currently investigating performance issue on system run=
ing above
>> btrs filesystem. Is it possible, that performance is impacted by lac=
k of
>> free space? Also, how to get info about real free space on btrfs vol=
ume?
>>
>> # btrfs-show /dev/sdb1
>> Label: opt =C2=A0uuid: 28a55827-e677-47a9-98d5-d31eb3d71436
>> =C2=A0 =C2=A0Total devices 1 FS bytes used 167.83GB
>> =C2=A0 =C2=A0devid =C2=A0 =C2=A01 size 240.00GB used *229.25GB* path=
 /dev/sdb1
>>
>> Btrfs Btrfs v0.19
>>
>> # btrfs filesystem df /opt
>> Data: total=3D213.23GB, used=3D165.26GB
>> System, DUP: total=3D8.00MB, used=3D40.00KB
>> System: total=3D4.00MB, used=3D0.00
>> Metadata, DUP: total=3D8.00GB, used=3D2.57GB
>>
>> # df -h /opt
>> Filesystem =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0Size =C2=A0Used =
Avail Use% Mounted on
>> /dev/sdb1 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 240G =C2=A0171G =
=C2=A0 59G =C2=A075% /opt
>>
>> How come that there is difference detween btrfs-show and df .. 40GB =
Is the
>> space really usead or can I claim it back? (there are no snapshots)
>>
>> # btrfs subvolume list /opt
>> #
>>
>
> The btrfs-show command is being deprecated. =C2=A0It's output can be =
easy
> to misunderstand, but it probably won't be corrected since it's going
> away at some point.

The output of "btrfs fi show /dev/whatever" is identical, and isn't
going away afaik.  That said, it is easy to misinterpret, although
that's probably unavoidable while still actually presenting that
information.
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" =
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2012-01-10 18:54 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-01-10 14:13 real free space on btrfs volume (performance impact) Michal Suba
2012-01-10 18:40 ` Mitch Harder
2012-01-10 18:54   ` cwillu

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.